Model reference · open weights

SraVaani-0.5-live

Audio ARTPARK-IISc Speech→text 1 build Open weights 732 dl/mo

SraVaani-0.5-live is an open-weight audio or speech model from ARTPARK-IISc. SraVaani-0.5-live (BF16) weighs 0 MB; the smallest configuration that runs it is RTX 3060 12 GB.

  • SraVaani-0.5-live is a multilingual automatic speech recognition model developed by ARTPARK-IISc that supports 63 Indian languages and dialects.
  • It utilizes a cache-aware FastConformer encoder with a CTC decoding head, exported as a TorchScript graph for low-latency streaming inference.
  • The model is licensed under the MIT license and offers four runtime latency settings ranging from 0 ms to 1040 ms lookahead.

Summary of the ARTPARK-IISc/SraVaani-0.5-live model card, 2026-10-01

What it is

Released byARTPARK-IISc
Released2026-07-16
VRAM0 MB for the weights

What it runs on

Memory and cards for SraVaani-0.5-live (BF16)

0 MBweights, file size
1.6 GBruntime overhead
CardOne streamMemory
RTX 3060 12 GBfits11.6 GB
RTX 4060 Ti 16 GBfits15.4 GB
RTX 3090 24 GBfits23.4 GB
RTX 4090 24 GBfits23.4 GB
RTX 5090 32 GBfits31.0 GB
L40S 48 GBfits44.0 GB
A100 80 GBfits78.2 GB
H100 80 GBfits78.1 GB
RTX PRO 6000 Blackwell 96 GBfits93.8 GB
DGX Spark (GB10) 128 GB unifiedfits107 GB
H200 141 GBfits138 GB
B200 180 GBfits176 GB

How it works

How audio & music work

Audio or textinputAudio modelrecognise / synthesiseText or audiooutputSpeech-to-text turns audio into text; text-to-speech and music models turn text into audio.
© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms